Perspectives in Molecular Imaging Through Translational Research, Human Medicine, and Veterinary Medicine Clifford R. Berry, DVM, DACVR,* and Predeep Garg, PhD† The concept of molecular imaging has taken off over the past 15 years to the point of the renaming of the Society of Nuclear Medicine (Society of Nuclear Medicine and Molecular Imaging) and Journals (European Journal of Nuclear Medicine and Molecular Imaging) and offering of medical fellowships specific to this area of study. Molecular imaging has always been at the core of functional imaging related to nuclear medicine. Even before the phrase molecular imaging came into vogue, radionuclides and radiopharmaceuticals were developed that targeted select physiological processes, proteins, receptor analogs, antibody-antigen interactions, metabolites and specific metabolic pathways. In addition, with the advent of genomic imaging, targeted genomic therapy, and theranostics, a number of novel radiopharmaceuticals for the detection and therapy of specific tumor types based on unique biological and cellular properties of the tumor itself have been realized. However, molecular imaging and therapeutics as well as the concept of theranostics are yet to be fully realized. The purpose of this review article is to present an overview of the translational approaches to targeted molecular imaging with application to some naturally occurring animal models of human disease. Semin Nucl Med 44:66-75 C 2014 Elsevier Inc. All rights reserved.

T

he advancement of nuclear medicine in veterinary medicine has paralleled most of the usage in humans, except in the area of myocardial imaging, as myocardial infarction secondary to primary atherosclerosis is a rare event in veterinary patients, particularly dogs and cats. However, planar scintigraphy (bone, renal, and thyroid scans) has remained a clinical tool used in a variety of orthopedic, renal, and thyroid disorders in veterinary patients.1 As PET has become a primary diagnostic technique and an accepted method for the evaluation of the human oncology patient, it seems that planar imaging has become a secondary player.2,3 Due to the costs involved in equipment, personnel, and infrastructure, PET, PET/CT, cyclotrons, and other PET techniques have not

*Department of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Florida, Gainesville, FL. †Department of Radiology, School of Medicine, Wake Forest University, Winston-Salem, NC. Address reprint requests to Clifford R. Berry, DVM, DACVR, Department of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Florida, 2015 SW 16th Ave., Gainesville, FL 32610. E-mail: [email protected]

66

0001-2998/14/$-see front matter & 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.semnuclmed.2013.10.002

become a widespread primary imaging modality in veterinary medicine.4 However, translational medicine and the concept of “one health” amalgamation has provided a forum for continued research in the area of spontaneous disease in veterinary patients as it relates to human disease. Although these terms have a variety of meanings depending on context, the current environment provides for the exploration of spontaneous diseases in animals.5 Starting with some basic definitions, translational medicine would be considered the science of taking the basic bench research to the patient and back. The bigger picture of translational medicine implies an interdisciplinary approach for understanding the pathobiology of a disease process based on the current understanding of the physiology of the cells involved (host's normal and abnormal cells) and how to then target specific therapies to these processes so that health can be restored to the affected individual. In the concept of one health or one medicine, we have purposefully used the term “amalgamation” to solidify the concept of union and unification between human and veterinary medicine.5 There are definitely differences. However, there are strong similarities in many of the spontaneous diseases and specific tumor types that occur in animals and their human counterparts.6

Perspectives in molecular imaging Certain animal vectors (bird species in particular) have been used as early indicators of new viral strains; whereas other diseases, specifically tumors, have been shown to be models of human tumors, such as canine osteosarcoma of the appendicular skeleton.6 The purpose of this review article is to look at radiopharmaceutical and other molecular imaging techniques that have been used for the evaluation of human and veterinary patients in a research or clinical context. Three areas that are discussed include a historical overview, current concepts of targeted imaging and therapy, and finally, an overview of spontaneously occurring animal models of human disease.

Historical Survey and Antibody Imaging Early nuclear medicine studies focused on spontaneous canine tumors6 and uptake of amino acids using N-13 as the radionuclide.7 Although a small number of dogs were used in this study, a wide variety of tumor types had an affinity for active accumulation (and presumed incorporation) of the radiolabeled amino acid L-glutamate for active protein synthesis in a hypermetabolic tumor.7 The search for an ideal imaging agent for a specific tumor type initially led to the use of monoclonal antibodies and antibody fragments. The early techniques for antibody formation described in 1975 provided the groundwork for antibody development and labeling.8 Tumor-associated antigens became an immediate antibody-antigen target that resulted in the development of a number of radiolabeled antibodies, Fab and F(ab′)2, diabodies, and single-chain variable fragments.9 Although antibodies have the ability for selective binding and the potential for a high sensitivity and specificity for specific tumor protein (typically located at the level of the cell membrane), cell receptor, or other form of antigen, their utilization has been limited because of slow blood pool clearance, long retention in normal tissues, nonspecific binding to a number of different tumor-associated antigens found in normal tissues, and the complex process of labeling and isolation before patient injection.9 In addition, intratumoral factors related to antibody-labeled radionuclide delivery include blood supply, tumor type, cellular microenvironment, and target antigen specificity with antigen expression levels associated with the tumor cells (cell membrane antigen turnover). Over time, other tumor proteins (peptides), small molecular ligands and synthetic graft copolymers have been developed and utilized.9 A number of Food and Drug Administration–approved IgG antibodies and Fab and F(ab′)2 radiopharmaceuticals are available for current use as primary radiopharmaceuticals for tumor identification, as well as diabodies and single-chain variable fragments using 99m Tc, 111In, and 123I as the radionuclides for single-photon emission computed tomography imaging are in preclinical trials pending approval by the Food and Drug Administration.9 In the case of PET radiolabeled antibody imaging, 18F is not an ideal radionuclide because of its short physical half-life;

67 however, other radionuclides of importance for antibody conjugation could include 68Ga, 124I, 64Cu, and 89Zr. In veterinary medicine, one of the first antibody-labeled techniques developed was used for the noninvasive imaging diagnosis of active heartworm infestation (Dirofiliaria immitis) in dogs.10 In this study, a monoclonal antibody was successfully labeled using a diethylenetriaminepentaacetic acid chelation technique. However, this technique and the Fab or F(ab′)2 follow-up studies were not developed into a routine imaging application although this early research cleared the way for development of blood kits used today for plasma detection of canine heartworm antigen. In later studies, nonspecific immunoglobulin accumulation was evaluated in a viral arthritis goat model.11 Nonspecific radiolabeled leukocyte imaging has also used for imaging the feline pancreas and detecting pancreatitis in cats.12,13 Going one step further, other clinical applications of nonspecific inflammatory techniques for imaging in small animals have included 111In-labeled transferrin for imaging protein-losing enteropathies in dogs as well as 111In-chloride, which was used in the evaluation of a canine osteomyelitis model.14,15 A number of dog models have been used for antibody labels using planar and single-photon emission computed tomography scintigraphy as well as PET imaging. For myocardial imaging, an instant kit method for Fab ′-labeled 99mTc was developed and tested in variety of clinical and research applications.16 Dogs with experimentally induced myocardial infarctions were imaged and showed equivalent levels of uptake compared with 111 In-antimyosin at the area of infarction. At the same time, 18F-labeled antimyosin monoclonal antibody fragments had active accumulation and uptake was seen in areas of damaged myocardium and ischemia with specific accumulation in a subendocardial position; however, owing to the sustained blood pool radioactivity, subtraction techniques using (15O) carbon monoxide were suggested by the authors.17 A sympathetic analog PET tracer was also developed that was tested in a canine myocardial arterial occlusion model where it was shown that (18F)-parafluorobenzyl guanidine uptake was delayed even after myocardial perfusion (imaged using 13 NH3) was reestablished.18 Fab fragment labeled with 18F was used in a group of dogs with naturally occurring osteosarcoma, a model for human appendicular skeletal osteosarcoma.19 In this study, 18 F-labeled Fab fragment of TP-3, a monoclonal antibody for human and canine osteosarcoma, had a biphasic blood clearance with a short half time and rapid radiopharmaceutical uptake in the primary osteogenic sarcoma site; 3 dogs with metastatic disease also had increased uptake. In a different study using 99mTc-methylene diphosphonate, the patterns of uptake in 25 dogs with spontaneously occurring osteogenic sarcoma showed that when there was a large tumor area and higher tumor radioactivity, the dogs were more likely to have early metastatic disease showing the predictive utility of bone scintigraphy.20 In a similar recent human study, these data were affirmed when it was shown that high (18F)-FDG maximum standardized uptake value numbers before and

68 after chemotherapy in 31 human patients with osteosarcoma were significantly correlated with a poor prognoses with respect to progression-free survival, overall survival, as well as overall tumor necrosis within the actual primary tumor at histology.21

Targeted Diagnostic Molecular Imaging Approaches—Beyond Antibodies At the core of molecular imaging is the ability to image noninvasively and characterize cellular and subcellular processes within human and veterinary patients.22 The merging of biology, imaging, biochemistry, and pharmacology has been the underlying mechanism of action for targeted nuclear medicine imaging since the beginning.22 Over time, the use of specific radiopharmaceuticals for evaluation of receptor tumor expression, metabolic pathways, and processes has come to selective molecular imaging related to targeting a specific protein, receptor, cytokine, tumor-related protein, process, metabolic pathway, and a specific nuclear or cytosolic event that might be expressed in neoplastic transformed cells.23-25 The primary broad areas of research have included cancer, myocardial and vascular disorders, other neurologic disorders with a metabolic basis, as well as normal physiological processes such as senility and programmed cell death. As an example, the complexities of cancer formation are just now being established.23-25 Fundamental aspects of cancer biology have come to light recently, but in truth, all of the pieces of the puzzle are not understood and the complex interactions of multiple signals result in difficulties when trying to image targeted pathways or developing specific treatments. However, certain areas of targeted therapy do show some degree of promise. It appears that the old adage, “the more we know, the more we don't know” applies in this area of research, particularly today. This only means that we do not have all the pieces of the puzzle but should not stop.

Cancer Molecular Imaging The reader is referred to an excellent review by Michalski and Chen25 given in 2011 for an overview of molecular imaging in cancer therapy. From a personalized approach or using the basic concepts of theranostics; patient treatment should be individualized based on clinical testing and imaging studies.26 These would form the basis of a decision tree for strategically placed therapy based on prior outcomes.26 Ideally, a molecular imaging study would provide multiple levels of information depending on the stage of imaging for the cancer patient. Therefore, a single molecular imaging study would not provide all the answers. The “quest” for a silver bullet study is not the real answer; however, an approach where a given molecular imaging study provides a directed answer in each of the following scenarios becomes the goal. Initially, the molecular imaging study would identify those patients with the specific tumor type with a high sensitivity and specificity so that the general population can be surveyed. This would provide for

C.R. Berry and P. Garg early cancer detection so that the appropriate therapeutic intervention can then be provided if necessary.26 However, this would go beyond just the basics of anatomical imaging (x-ray, CT, and routine MR), but should be evidence-based for a given tumor type; this implies that histologic conformation of the tumor with staging based on pathology and the initial molecular imaging study. Next, there would be a targeted molecular imaging study that provides individualized information about the type of cancer that is involved. This is different in that the molecular imaging study would assess the primary and metastatic (regional and systemic) sites and would be quantified in a semi-automated or manual fashion. Additionally, the molecular changes identified should provide some prediction of therapeutic susceptibility as well as the potential for tumor heterogeneity and other important factors for the prediction of local invasion and ultimately systemic spread of the cancer cells.27 At this stage, based on the results of these tests or imaging studies, treatment paradigm would provide relevant information that could be used for tumor targeted therapy, thereby taking advantage of the tumor specifics and the known local tumor environments for tumor control before metastatic disease. This would include better treatment delivery systems that would target specific biological properties or processes related to the particular tumor based on the molecular imaging principles that have been previously described. Radiopharmaceutical development and the development of systemic radiotherapy or chemotherapies or both could also be created that would allow for active accumulation and retention of beta or alpha emitters or specific cytotoxic drugs for directed tumor cell kill. Once treatment has been implemented, there should be molecular imaging studies that are again personalized to evaluate tumor response (or nonresponse) as well as local and systemic control of the tumor itself. Each of these imaging techniques has to go beyond standardized Response Evaluation Criteria in Solid Tumors anatomical imaging protocols using measured data and should provide prognostic indicators of tumor progression toward cellular arrest and cell death.27,28 Metabolic (glucose) imaging has been the primary strength of PET imaging with 18F-FDG being the primary radiopharmaceutical used.22 This imaging modality has been used for differentiating malignant and benign lesions and documenting metastatic lesions within regional lymph nodes, metastatic bone lesions, as well as those within the lungs.29 In addition, the other primary area that 18F-FDG-PET has been used in is follow-up studies after therapy to document progression or regression of metastatic disease.25 An estimated 19% of current PET studies performed in the United States are to monitor patient response to therapy.25 However, an initial inflammatory reaction associated with tumor cell kill can increase 18F-FDG uptake at the site of the primary or metastatic tumor. All this being said, a review of the current targets of the cancer cell progression is appropriate with a look at molecular imaging techniques that have been targeted at specific tumor cell stages or characteristic hallmark features of tumor biology. A recent review by Hanahan and Weinberg23 updated their original description for these hallmark changes that occur

Perspectives in molecular imaging within the tumor cell. Neoplastic transformation from a normal differentiated cell to a tumor cell has a series of changes that have been expanded to include 8 biological hallmarks. These biological changes that occur within the cancer cell as it replicates and influences the surrounding microenvironment were originally described as 6 hallmark changes with 2 broader “enabling characteristics” that are incorporated during the formation of the macroscopic tumor from the initial monoclonal tumor cell.23 These hallmark features are not necessarily linear, instead they presumably occur in somewhat of a randomized order, as the mutations necessary for change are random. All of these changes are characteristic of cancer cells and are required for allowing sustained cancer cell growth and metastasis. The 6 hallmarks originally described include continual cellular proliferation and proliferative signaling, evasion of normal growth suppression, activation of invasive cellular attributes and metastasis, entrance into a replicative immortality, induction of a blood supply for sustained growth, and the resistance of normal cellular programmed death via the mechanisms of apoptosis or cellular senility.23 The 2 newer concepts of cancer cell features include “enabling” hallmarks of avoiding immune-mediated destruction (lack of recognition as a “foreign” cell and thereby avoiding destruction by B and T lymphocytes, natural T-killer cells, and macrophages) as well as the deregulation of cellular energetics so that neoplastic proliferation can be sustained.23 In addition, there is current literature and understanding that provides evidence for tumorinduced inflammation that results in promotion of natural growth of the tumor itself.23 Inherent to some of these cellular properties is the progression of genomic instability secondary to progressive mutations. This provides (along with the support from stromal cells) for tumor heterogeneity, even though the original tumor is derived from a single cell (clonal expansion from a nonlethal genetically altered cell).30 If one looks at each cancer cell hallmark, one can see several obvious dynamics and make several observations about what we are doing now and how to potentially approach future aspects of molecular imaging. The first is that there would not be a “magic bullet” approach to all cancers or even groups of cancers.31 The approach should include both novel molecular imaging techniques (that fundamentally are going to have to be a multimodality approach) and a combination of therapeutic strategies specific to a group of tumors or certain tumor type.32-53 This is particularly true for any tumors that progress quickly, are invasive on local and systemic levels, develop resistance to chemotherapy and radiation rapidly, and cannot be surgically excised without radical resection. It should also be clear that blocking a single pathway (or hallmark as described by Hanahan and Weinberg) with a given therapeutic approach is not going to ultimately control the cancer, as there are many interconnected pathways already understood or yet to be determined that provide cross talk and signaling so that a given pathway is not a required source of “life” for the tumor cell. A multidisciplinary approach that attacks the cancer at multiple “hallmark biological concepts” is required. In addition, because of this, the molecular imaging approach would be such that different facets of these hallmarks of a specific tumor need to be imaged. This would require the emergence of a wider number

69 of molecular imaging approaches that include multiple imaging modalities and target the different levels of imaging described earlier.32-53 To this end, nuclear imaging cannot be seen as a competing process (with contrast-enhanced ultrasound, optical imaging, and contrast-enhanced computed tomography or magnetic resonance imaging) but must play a complimentary role. The basis for molecular imaging, by extension, implies a dynamic process such that input factors (vascularity and any active, relevant uptake kinetics) as well as retention or washout factors should be a part of the imaging process.49,52,53 In addition, radiolabeled delivery agents such as liposomes should be explored for novel drug and radiotherapy strategies.35-37 The features of kinetic physiological imaging are by their very nature a given strength of nuclear medicine, whether using list-mode or dynamic frame-mode planar imaging or advanced PET kinetic modeling. “Hot spot” imaging, even though this can and should be a component of the interpretation process, cannot be the only endpoint for a molecular imaging study. The capacity of tumors to proliferate independently of external stimulus and other paracrine influences is felt to be secondary to an acquired cellular autonomy within the tumor cell provided by oncogenes.23 Proto-oncogenes, present in all somatic cells, undergo mutations that, if not repaired, result in a tumor cell with an active oncogene(s) that promotes oncoprotein(s) synthesis. These oncoproteins are no longer regulated by normal nuclear or cytosolic cellular growth factors (either promoters or inhibitors). Thereby, the cells become autonomous and develop sustained proliferative capabilities. Focusing on only imaging cellular proliferation neglects the potential targets that might include specific cell membrane receptors regulating (promoting or inhibiting) cell proliferation, signal transduction proteins of the cell membrane, second messenger cytosolic pathways, and proteins and cell transcription factors important to each of the oncoproteins involved in this process. Some of these mutations of the proto-oncogenes into oncogenes also set up a self-regulatory autocrine loop for cell proliferation by producing specific growth factor(s) or growth factor promoter(s) or agonists. Overexpression of these oncogene-related growth factors and growth factor cell membrane receptors are found in a number of different tumor types. These factors include platelet-derived growth factor (as seen in glioblastomas), transformation growth factor α (as seen in soft tissue sarcomas), as well as proteins involved in signal transduction (ras signal transduction point mutations in melanomas).23 Pigment epithelium–derived factor is a known serine protease inhibitor (or serpin) that has antiangiogenic (major antagonist to vascular endothelial growth factor or VEGF), antitumorigenic, and antimetastatic activities.54 Reporter genes are genetic markers that encode for readily detected proteins and receptors downstream from the nucleus and typically are located in the cytosol. This area of imaging has been called molecular-genetic imaging and allows incorporation of a “reporter” into the genetic cellular code of a “transformed” tumor cell that can then be monitored in some form so that one knows when oncogenes or more specifically downstream oncoproteins are expressed.

70 Growth factor receptors that have been transcribed from oncogenes undergo automated stimulation via a dimerization process that results in continuous activation of a tyrosine kinase–dependent signaling cascade within the cytosol. This results in continuous mitogenic stimulation within the tumor cell. Targeted therapy at these overexpressed and stimulated tyrosine kinase receptors has been taken advantage of in several tumors so as to slow down the growth or expansion of the tumor itself. The ras point mutations are the most common abnormality of proto-oncogenes in human tumors. Point mutations of the ras genome result in an activated ras protein that remains in its activated form with decreased guanosine triphosphatase activity. This results in the inability of guanosine triphosphate to be hydrolyzed and to return to the inactive guanosine diphosphate state. This stimulates regulators of cell proliferation such as mitogen-activated protein kinase cascade, which activates transcription and a continuous cell cycle proliferative state. However, in canine and feline tumors, only canine melanoma has been shown to have a Ha-ras point mutation.55 In a mouse tumor model, 212Pb-DOTA-AE1 was used as a therapeutic agent that targeted a human ovarian cell line that expressed the HER-2/neu oncoprotein.56 Several studies have shown a correlation between 18F-FDG uptake relative to c-erbB-2 oncoproteins (HER-2).57,58 In a study, the proto-oncogene, RET (rearranged during transfection protein) was targeted because of the multiple potential intracellular signaling cascades that are involved in cell cycle regulation.59 In this study, vandetanib (kinase inhibitor of a variety of cell kinase receptors including VEGF receptor, epidermal growth factor receptor, and RET-tyrosine kinase) resulted in a decrease or downregulation of glucose, dihydroxyphenylalanine, and thymidine metabolism in a thyroid carcinoma mouse model using micro-PET and 18F-FDG standardized uptake ratios. Mutations in the cyclins, nonreceptor tyrosine kinases, cyclin-dependent kinases (CDK), and CDK inhibitors result in deregulation of normal cell cycle checkpoints, which would result in overexpression of some of the CDKs.23 The result is a breakdown in normal checkpoints between G1 to synthesis and G2 to mitosis within the cell cycle and continuous cell proliferation. Additionally, mutations of the inhibitors of the CDKs that would render these proteins inactive would also result in the overexpression of CDKs. Thereby, the initiation process of mitosis (DNA replication) is controlled by the activity of the specific cyclin E-CDK2.23 There are tumor suppressor genes that can undergo mutation and thereby do not suppress normal cell cycle arrest and apoptosis in response to DNA damage. The most well known of these genes is the p53 tumor suppressor gene.23 Homozygous loss of the p53 gene occurs in virtually every type of human cancer, including pulmonary, colon, and breast carcinomas.23 Tumor suppressor genes transcribe a series of proteins that stop cell proliferation and thereby uncontrolled growth. These genes normally recognize any form of DNA damage that should be repaired and stops the cell from entering into the cell cycle and proliferating. These suppressor genes produce proteins that regulate cellular progression toward senescence or apoptosis. The functions of p53 to suppress neoplastic transformation include activation of

C.R. Berry and P. Garg temporary cell cycle arrest (quiescence) for DNA repair, induction of permanent cell cycle arrest (senescence), or initiation of programmed cell death (apoptosis). In a breast cancer mouse model, it has been shown that an increased 18 F-FDG at the site of the tumor is due to inflammation associated with chemotherapy rather than increased tumor glycogenesis because there is a decrease in uptake of 3′-18Ffluoro-3′-deoxy-L-thymidine.60 The later radiotracer is cell cycle dependent and actively phosphorylated by cytosolic thymidine kinase-1. This leads to intracellular trapping of the radiotracer and active accumulation is assumed to reflect active cell proliferation. Inflammatory cells tend not to actively proliferate at the site of inflammation so that active tumor uptake after chemotherapy or radiation therapy would be related directly to tumor proliferation and thereby more accurately reflect response to therapy.60,61 However, limitations, of 3′-18F-fluoro-3′-deoxy-L-thymidine in human studies have been found including sensitivity and specificity differences with different tumor types, high bone marrow and liver background radioactivity, and false positives in areas of lymphocyte proliferation such as reactive lymph nodes.25 Lack of progression of a normal aging cell toward programmed death or apoptosis can result from any combination of factors including activation of growth-promoting oncogenes, inactivation of growth-suppressing tumor suppressor genes, and mutations in the genes responsible for normal regulation of apoptosis.23 Normally, significant genomic insult should result in progression toward programmed cell death by apoptosis so as to prevent the possible transmission of any significant mutation such as progression toward a malignant tumor cell. Other potential triggers for apoptosis include loss of cell-to-cell contact (paracrine secretory signal mechanisms) or loss of normal basement membrane structural support of cells.23 Apoptosis is regulated by either intrinsic (cellular injury —growth factor withdrawal, DNA damage, or protein abnormalities) or extrinsic factors or pathways (receptor ligand interactions such as tumor necrosis factor receptor or Fas receptors).23 In either case, specific enzymes called caspases are activated resulting in cytoskeletal degradation. Additionally, activation of endonucleases results in apoptotic body formation with ensuing macrophage phagocytosis of the degraded membrane-bound cellular debris. To avoid these natural events scheduled for typical programmed cell death, mutations can result in the upregulation and overexpression of specific apoptosis inhibitory proteins called inhibitors of apoptosis proteins.23 In addition, mutations in the p53 genetic code can result in dysfunction of the proapoptotic effects of normally transcribed proteins that occur in the event of lethal DNA damage. It has been shown that apoptosis is the most common mechanism for treatment-induced tumor cell death, hence the lack of apoptosis can possibly be used as a sign of treatment lack of response.62 Imaging of apoptosis via external pathways that result in changes in the cell membrane composition has been explored. This includes using reliable cell membrane markers of apoptotic cell death such as phosphatidylethanolamine and phosphatidylserine. These markers aid in normal cell degradation and enhance phagocytosis of apoptotic bodies.

Perspectives in molecular imaging Annexin V, a human anticoagulant protein, has been shown to bind to phosphatidylserine and has been used for labeling apoptotic cells.62-67 Blankenburg et al63 have shown increased uptake in areas of lymphoma after cyclophosphamide treatment using 99mTc-annexin V in a mouse xenograft model. Over time, different variations of annexin V have been used where the amino acid terminal has provided a mechanism for chelation of the 99mTc and decreased biliary and renal excretion, thereby allowing better visualization of tumors in the area of the abdomen.65 Cellular senility can be described as the cell's inability to enter into mitosis and divide. This appears to be driven by telomere shortening over time as the cell ages.68-70 In tumor cells, the ability to avoid cell senescence and any type of mitotic crisis (based on chromosomal aberrations such as dicentric chromosomes and double strand DNA breaks), is based primarily on the reactivation of telomerase, a nuclear enzyme active in normal stem cells. It has been shown that telomerase reverse transcriptase is present in most malignant cells (485%) in human medicine but is not detectable in normal somatic cells. In a study, an antisense oligonucleotide that was radiolabeled so as to noninvasively measure telomerase reverse transcriptase messenger RNA was shown to actively accumulate in vivo in a mouse model using mammary tumor cells.71 A necessary requirement for all malignant tumors is the production of cellular signals to stimulate constructs of the “ideal” microenvironment for tumor cell survival. The tumor cell would need to stimulate angiogenesis, tumor cellular invasion and metastasis and the development and stimulation of the appropriate stromal microenvironment.23,72-76 Because tumors require adaptation of their microenvironment, there are a number of imaging modalities that have taken advantage of induced angiogenesis of malignancy. These modalities include contrast-enhanced CT, ultrasound, and MR as well as dynamic imaging studies using planar scintigraphy and PET.75 There is a zone of 1-2 mm in dimension from which tumor cells must have an extracellular matrix that allows for vascular delivery of oxygen and nutrients to the tumor cells and the removal of waste products from the tumor cells. Tumor cells have various induction mechanisms for either angiogenesis (induction of new vessels from previously existing capillary beds) or vasculogenesis, where hormonal influences recruit endothelial cells from the bone marrow directly to the tumor microenvironment for the formation of new blood vessels. As seen with the other cancer hallmarks, there is a complex interaction between angiogenic promoters and inhibitors. Some of these promoters may in fact be derived from paracrine interactions with the inflammatory response (macrophages) or other stromal cells recruited by the tumor for establishing its own “ideal” microenvironment. Proteases (such as basic fibroblastic growth factor [bFGF]) can result in proangiogenic stimulation in a region of tumor development.23 A multimodality approach to the imaging of tumor angiogenesis was described by Cai and Chen. In this approach, the use of nanoparticles was suggested as an ideal way to target neovascularization associated with tumor angiogenesis.75 A significant advantage of nanoparticles over labeled VEGF or integrin αvβ3 is that the nanoparticle can be multifunctional

71 at the same time. It can be tagged with an imaging label, a target ligand, as well as a therapeutic drug.75 Relative cell hypoxia is a signal for the turning on of angiogenic recruitment.77-82 Proangiogenic cytokines that are normally produced in response to hypoxia include VEGF and bFGF. These cytokines create an angiogenic gradient such that the ingrowth of vessels and endothelial cellular migration is toward the tumor cells. Again, p53 is important in angiogenesis from the stand point that loss of p53 in tumor cells results in the expression of proangiogenic proteins such as VEGF as well as repressing the antiangiogenic cytokine thombospondin-1. Elevated plasma and urinary levels of VEGF and bFGF have been found in a number of different human tumors resulting in a targeted molecular approach to these tumors using bevacizumab, an anti-VEGF antibody, as a potential adjuvant therapy in the treatment of the tumors that overexpress VEGF.23,73,83-89 Tumor cell invasion involves direct paracrine cell-cell interactions and loss of normal interactions and intercellular proteins such that there is degradation and breakdown of these support proteins within the extracellular space. Novel extracellular matrix proteins are expressed by the tumor cells (autocrine chemotactic migration), resulting in a “global positioning system” self-made guidance system for the tumor cell to invade the extracellular space.90 As most tumors have some degree of hypoxia within them, specific targets for hypoxic imaging have also been evaluated.77-82 Hypoxic factors that contribute include sparse arteriolar vasculature, high oxygen consumption rate, low vascular density, disorganized vascular arrangement with arteriovenous shunts, and loss of red cell oxygen carrying capacity because of the viscosity of the blood and relative stiffness of the red blood cell membrane.80-82 It has been shown that patients with tumors that have a relatively high hypoxic environment and are surviving under oxidative stresses often have a poorer prognosis and decreased survival times. 18F-PET agents using the base of a 2-nitroimidazole have been developed and evaluated in a number of human cancers.80-82 The use of these radiotracers suggests that patients with large components of tumor hypoxia would benefit from targeted types of treatment including intensity-modulated radiation therapy, hypoxic cytotoxins, and hypoxia-inducible factor-1 inhibitors. The later drugs block the hypoxiainducible factor-1 upregulation of genes and thereby the proteins that have protectant effect for cells under oxidative stress or hypoxia. Within the nucleus of the tumor, the basis for mutagenesis is DNA damage, replication error, or actual chromosomal aberrations, such as deletions, ring chromosome formation, point mutations, chromosomal crossing over, abnormal karyotypic expressions, hybrid fusion of chromosomes, and abnormal chimeric protein production resulting in chromosomal bridging, fusion, and breakage. Other nuclear changes include microRNA (miRNA) alterations that result in lack of normal posttranscriptional cytoplasmic silencing of gene expression. This could result in the overexpression of a given oncoprotein if the miRNA is reduced or a reduction in a tumor suppressor protein if there is overproduction of the miRNA and increased translational repression. Molecular imaging of

C.R. Berry and P. Garg

72 miRNAs has recently been reviewed.91 This class of noncoding RNAs has been shown to be important in the translational regulation of messenger RNAs that are involved in cell proliferation, differentiation, and development as well as apoptosis. Different miRNAs have been shown to be both tumor suppressive as well as having the potential to be oncogenic in nature. Two approaches currently used to image miRNAs include reporter genes and fluorescent beacon imaging techniques.91 However, these techniques have not been developed for human imaging studies. Two other nonspecific imaging PET-related targets that have been developed include choline receptors, which are upregulated within the cell membrane of tumor cells,92-95 as well as glutamine metabolism in tumor cells.96,97 As choline transporter cell membrane protein and intracellular choline kinase are overexpressed in malignant tumors, labeled choline radionuclides (including 11C and 18F) have been synthesized for PET. Specific tumors of the central nervous system, prostate gland, breast, and esophagus have been shown to have avidity for 18F-choline uptake. In addition, derivatives of choline (fluoromethylated analog of choline) have been successfully labeled using 18F. The trapping and phosphorylation of choline by choline kinase within the tumor cell has been correlated with mitogenic activity or as a measure of cell proliferation. Additionally, the use of magnetic resonance spectroscopy of in vivo tumors has shown an elevated level of phosphocholine, which has been correlated with malignant transformation, tumor growth rate, metastatic potential, and therapeutic response.92 Because of the Warburg effect, it is known that tumor cells convert from normal glucose metabolism using the Krebs cycle to sustained glycolysis with resulting high levels of lactic acid. Alternative metabolic strategies of cell survival have shown that glutamine is an important tumor cell nutrient and is a contributor to the metabolic success of proliferating tumor cells. Upregulation of glutamine transporter (cell membrane channel), glutaminase (cytosolic enzyme with conversion to glutamate which can then enter the Krebs cycle after conversion to α-ketoglutarate) and the myc oncogene involved in the increase of cellular processing of glutamine has been shown to occur in tumor cells.96 Recently, a 11C-labeled glutamine has been synthesized and studied in vitro as well as in vivo using mouse tumor model.97 This study showed that glutamine metabolism in 2 different neural tumor cell lines (glioma and astrocytoma) was elevated with increased uptake in the tumor as well as high cardiac and pancreatic uptake.

Spontaneous Tumors in Domestic Animals The limitations of most of the cited studies are the use of small animal (rodent) models where tumor cells are introduced. In some cases, spontaneous tumors in genetically engineered mice can be evaluated but the development times are variable and the endpoints are harder to control. However, small animal (murine) PET/CT scanners have been introduced and instrumentation adapted for high-resolution imaging of these

animal models.98,99 The use of spontaneous tumors of domestic animals as models of human disease has been well established for certain tumor types.5,100-104 These include canine glioma, osteogenic sarcoma, lymphoma (non-Hodgkin's lymphoma), malignant melanoma (not associated with sun carcinogenesis), and soft tissue sarcomas.5 There are several areas where spontaneous tumors in dogs and cats are attractive as models for human tumor counterparts. The first is that they share common environments, hence any carcinogenic environmental exposure would be similar. They are a more natural outbred population of animals when compared with inbred strains of specific mice and rat species. The malignancies develop spontaneously and have a heterogeneous presentation, more representative of presentation of humans with spontaneous tumors. There are certain tumor types, such as osteogenic sarcoma in dogs and non-Hodgkin's lymphoma in cats that present with a higher spontaneous incidence than in humans, therefore they provide a large population for study. The malignancies develop and spread in a rapid fashion; consequently, long study designs are not necessary with most clinical trial endpoints being reached within 12-24 months. By virtue of the size of dogs and cats compared with the murine species, the collection of body samples (fluids, blood, cerebrospinal fluid, urine, etc.) and the use of routine imaging modalities without specialized equipment are possible.106 All this being said, there is a need to take advantage of these models as part of large-scale, multiinstitutional studies.

Summary Currently, it is estimated that there is a 30% failure rate of PET 18 F-FDG imaging studies for the detection of human cancers and metastatic lesions.97-106 Molecular imaging is at the forefront for diagnosing the tumor, understanding the pathogenesis of local and metastatic tumors, as well as evaluating the tumor response to therapies that are used to treat the tumor itself. Additionally, these investigations provide the framework for use of radiolabeled therapies using alpha emitters and other novel radionuclides for treatment.107,108 Molecular imaging has to provide high sensitivity and specificity for the molecular process that it is evaluating. In addition, the kinetics of the process are important for the evaluation of prognosis and treatment response and ultimate outcomes for the tumor evaluated. Animal models of tumor biology have been used and will continue to be used for the evaluation of radiopharmaceutical development. The concept of “one health” promotes this use so that predictors of outcomes in spontaneous disease in animals can become predictors of spontaneous disease in people. In this article, a brief overview of the different hallmarks of cancer and where different targets have been directed toward therapy and molecular imaging in the past have been discussed. To move forward toward theranostics, the implications of benchtop to bedside translation are onerous. However, in this process, we cannot forget that nations capable of supporting these infrastructures in their health care system will have the technologies. In veterinary

Perspectives in molecular imaging medicine, there are no third-party payments and regulators for what test is required or which test needs to be done next. In the pets that we serve, there is a population of dogs and cats that benefit from the benchtop research being done now as well as human clinical trials. More importantly, the question is, are the clinical trials and molecular imaging techniques being evaluated in spontaneous diseases in veterinary patients that are appropriate models for human disease so that we can better serve the concept of “one health” and people on a more global level.

References 1. Daniel GB, Berry CR, (eds): Textbook of Veterinary Nuclear Medicine (ed 2). Philadelphia, PA: American College of Veterinary Radiology; 2006 2. Vallabhajosula S: 18F-Labeled positron emission tomographic radiopharmaceuticals in oncology: An overview of radiochemistry and mechanisms of tumor localization. Semin Nucl Med 2007;37:400-419 3. Von Schulthess GK, Kuhn FP, Kaufmann P, et al: Clinical positron emission tomography/magnetic resonance imaging applications. Semin Nucl Med 2013;43:3-10 4. Enserink M: Initiative aims to merge animal and human health science to benefit both. Science: 2007;316(15):1553 5. Garner MB: Current information on feline and canine cancers and relationship or lack of relationship to human cancer. J Natl Cancer Inst 1971;46:281-290 6. Gelbard AS, Christie TR, Clarke LP, Laughlin JS: Imaging of spontaneous canine tumours with ammonia and L-glutamine labeled with N-13. J Nucl Med 1977;18:718-723 7. Kohler G, Milstein C: Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 1975;256:495-497 8. Kaur S, Venktaraman G, Jain M, et al: Mini-review: Recent trends in antibody-based oncologic imaging. Cancer Lett 2012;315:97-111 9. Kaur S, Venktaraman G, Jain M, et al: Recent trends in antibody-based oncologic imaging. Cancer Lett 2012;315:97-111 10. Daniel GB, Allhands RV: In vitro testing of 99mTc-labeled monoclonal antibodies specific for Dirofiliaria immitis. Vet Radiol 1988;29:31-36 11. Papageorges M, Gavin PR, Barbee DD, et al: Nonspecific accumulation of Tc-99m-immunoglobulins in chronic and acute arthritis. Vet Radiol 1992;33:55-61 12. Head LL, Daniel GB, Becker TJ, et al: Use of computed tomography and radiolabeled leukocytes in a cat with pancreatitis. Vet Radiol Ultrasound 2005;46:263-266 13. Head LL, Daniel GB, Tobias KM, et al: Evaluation of the feline pancreas using computed tomography and radiolabeled leucocytes. Vet Radiol Ultrasound 2003;44:420-428 14. Berry CR, Guilford WG, Koblik PD, et al: Scintigraphic evaluation of four dogs with protein-losing enteropathy using 111Indium-labeled transferrin. Vet Radiol Ultrasound 1997;38:221-225 15. Hoskinson JJ, Daniel GB, Patton CS: Indium-111-chloride and threephase bone scintigraphy: A comparison for imaging experimental osteomyelitis. J Nucl Med 1991;32:67-75 16. Pak KY, Nedelman MA, Kanke M, et al: An instant kit method for labeling antimyosin Fab′ with technetium-99m: Evaluation in an experimental myocardial infarct model. J Nucl Med 1992;33:144-149 17. Zalutsky MR, Garg PK, Johnson SH, et al: Fluorine-18-antimyosin monoclonal antibody fragments: Preliminary investigations in a canine myocardial infarct model. J Nucl Med 1992;33:575-580 18. Berry CR, Garg PK, DeGrado TR, et al: Para-[18F]fluorobenzylguanidine kinetics in a canine coronary artery occlusion model. J Nucl Cardiol 1996;3:119-129 19. Page RL, Garg PK, Garg S, et al: PET imaging of osteosarcoma in dogs using a fluourine-18-labeled monoclonal antibody Fab fragment. J Nucl Med 1994;35:1506-1513 20. Forrest LJ, Dodge RK, Page RL, et al: Relationship between quantitative tumor scintigraphy and time to metastasis in dogs with osteosarcoma. J Nucl Med 1992;33:1542-1547

73 21. Costellow CM, Macapinlac HA, Madewell JE, et al: 18F-FDG PET/CT as an indicator of progression-free and overall survival in osteosarcoma. J Nucl Med 2009;50:340-347 22. Phelps ME: PET: The merging of biology and imaging into molecular imaging. J Nucl Med 2000;41:661-681 23. Hanahan D, Weinberg RA: Hallmarks of cancer: The next generation. Cell 2011;144:646-674 24. Reddy S, Robinson MK: Immuno-positron emission tomography in cancer models. Semin Nucl Med 2010;40:182-189 25. Michalski MH, Chen X: Molecular imaging in cancer treatment. Eur J Nucl Med Mol Imaging 2011;38:358-377 26. DeNardo GL, DeNardo SL: Concepts, consequences and implications of theranosis. Semin Nucl Med 2012;42:147-150 27. Eisenhauer EA, Therasse P, Bogarts J, et al: New response evaluation criteria in solid tumors: Revised RECIST guideline (ver. 1.1). Eur J Cancer 2009;45:228-247 28. Van Persijn van Meerten EL, Gelderblom H, Bloem JL: RECIST revised: Implications for the radiologist. A review article on the modified RECIST guideline. Eur Radiol 2010;20:1456-1467 29. Lowe VJ, DeLong DM, Hoffman JM, et al: Optimum scanning protocol for FDG-PET evaluation of pulmonary malignancy. J Nucl Med 1995;36:883-887 30. Almendro V, Marusyk A, Polyak K: Cellular heterogeneity and molecular evolution in cancer. Annu Rev Pathol Mech Dis 2013;8:227-302 31. Baum RP, Kulkarni HR, Carreras C: Peptides and receptors in imageguided therapy: Theranostics for neuroendocrine neoplasms. Semin Nucl Med 2012;42:190-207 32. Craciunescu OI, Thrall DE, Vujaskovic Z, et al: Magnetic resonance imaging: A potential tool in assessing the addition of hyperthermia to neoadjuvant therapy in patients with locally advanced breast cancer. Int J Hyperthermia: 2010;26(7):625-637 33. Dewhirst MW, Thrall DE, Palmer G, et al: Utility of functional imaging in prediction or assessment of treatment response and prognosis following thermotherapy. Int J Hyperthermia 2010;26:283-293 34. Viglianti BL, Lora-Michiels M, Poulson JM, et al: Dynamic contrastenhanced magnetic resonance imaging as a predictor of clinical outcome in canine spontaneous soft tissue sarcomas treated with thermoradiotherapy. Clin Cancer Res 2009;15:4993-5001 35. Kleiter MM, Yu D, Mohammadian LA, et al: A tracer dose of technetium99m-labeled liposomes can estimate the effect of hyperthermia on intratumoral doxil extravasation. Clin Cancer Res 2006;12:6800-6807 36. Matteucci ML, Anyarambhatla G, Rosner G, et al: Hyperthermia increases accumulation of technetium-99m-labeled liposomes in feline sarcomas. Clin Cancer Res 2000;6:3648-3755 37. Matteucci ML, Thrall DE: The role of liposomes in drug deliver and diagnostic imaging: A review. Vet Radiol Ultrasound 2000;41: 100-107 38. Inaba Y, Lindner JR: Molecular imaging of disease with targeted contrast ultrasound imaging. Transl Res: 2011;159(3):140-148 39. Lawrence J, Rohren E, Provenzale J: PET/CT today and tomorrow in veterinary cancer diagnosis and monitoring: Fundamentals, early results and future perspectives. Vet Comp Oncol 2010;8:163-187 40. Kiessling F, Fokong S, Koczera P, et al: Ultrasound microbubbles for molecular diagnosis, therapy and theranostics. J Nucl Med 2012;53: 345-348 41. Hwang M, Lyshchik A, Fleischer AC: Molecular sonography with targeted microbubbles: Current investigations and potential applications. Ultrasound Q 2010;26:75-82 42. Hong Shan X, Hu H, Xiong F, et al: Targeting Glut1-overexpressing MDA-MB-231 cells with 2-deoxy-D-glucose modified SPIOs. Eur J Radiol 2012;81:95-99 43. Ahn SJ, Choi SH, Kim YJ, et al: Histogram analysis of apparent diffusion coefficient map of standard and high B-value diffusion MR imaging in head and neck squamous cell carcinoma: A correlation study with histological grade. Acad Radiol 2012;19:1233-1240 44. Stevenson JP, Rosen M, Sun W, et al: Phase I trial of the antivascular agent combretastatin A4 phosphate on a 5-day schedule to patients with cancer: Magnetic resonance imaging evidence for altered tumor blood flow. J Clin Oncol 2003;21:4428-4438

74 45. Furman-Haran E, Schechtman E, Kelcz F, et al: Magnetic resonance imaging reveals functional diversity of the vasculature in benign and malignant breast lesions. Cancer 2005;104:708-718 46. Dale BM, Jesberger JA, Lewin JS, et al: Determining and optimizing the precision of quantitative measurements of perfusion from dynamic contrast enhanced MRI. J Magn Reson Imaging 2003;18:575-584 47. Griebel J, Mayr NA, de Vries A, et al: Assessment of tumor microcirculation: A new role of dynamic contrast MR imaging. J Magn Reson Imaging 1997;7:111-119 48. Stramare R, Beltrame V, Gazzola M, et al: Imaging of soft tissue tumors. J Magn Reson Imaging 2012;37:791-804 49. Tofts PS: Modeling tracer kinetics in dynamic Gd-DTPA MR imaging. J Magn Reson Imaging 1997;7:91-101 50. Brasch R, Pham C, Shames D, et al: Assessing tumor angiogenesis using macromolecular MR imaging contrast media. J Magn Reson Imaging 1997;7:68-74 51. Eisenbrey JR, Forsberg F: Contrast-enhanced ultrasound for molecular imaging of angiogenesis. Eur J Nucl Med Mol Imaging: 2010;37(suppl 1):S138-S146 52. Reddick WE, Taylor JS, Fletcher BD: Dynamic MR Imaging (DEMRI) of microcirculation in bone sarcoma. J Magn Reson Imaging 1999;10: 277-285 53. Gore JC, Yankeelov TE, Peterson TE, et al: Molecular imaging without radiopharmaceuticals? J Nucl Med 2009;50:999-1007 54. Becerra SP, Notario V: The effects of PEDF on cancer biology: Mechanisms of action and therapeutic potential. Nat Rev Cancer 2013;13:258-271 55. Watzinger F, Mayr B, Gamerith R, et al: Comparative analysis of RAS proto-oncogene mutations in selected mammalian tumors. Mol Carcinog 2001;30:190-198 56. Horak E, Hartmann F, Garmestani K, et al: Radioimmunotherapy targeting of HER2/neu oncoprotein on ovarian tumor using lead-212DOTA-AE1. J Nucl Med 1997;38:1944-1950 57. Groheux D, Giacchetti S, Moretti JL, et al: Correlation of high 18F-FDG uptake to clinical, pathological and biological prognostic factors in breast cancer. Eur J Nucl Med Mol Imaging 2011;38:426-435 58. Seregni E, Coli A, Mazzucca N: Circulating tumuor markers in breast cancer. Eur J Nucl Med Mol Imaging: 2004;31(suppl 1):S15-S22 59. Walter MA, Benz MR, Hildebrandt IJ, et al: Metabolic imaging allows early prediction of response to vendatanib. J Nucl Med 2011;52: 231-240 60. Brepoels L, Stroobants S, Verhoef G, et al: 18F-FDG and 18F-FLT uptake early after cyclophosphamide and mTOR inhibition in an experimental lymphoma model. J Nucl Med 2009;50:1102-1109 61. Been LB, Suurmeijer AJH, Cobben DCP, et al: [18F]FLT-PET in oncology: Current status and oppurtunities. Eur J Med Mol Imaging 2004;31:1659-1672 62. Hersey P, Zhang XD, Mhaidat N: Overcoming resistance to apoptosis in cancer therapy. Adv Exp Med Biol 2008;615:105-126 63. Blankenberg FG, Katsikis PD, Tait JF, et al: Imaging of apoptosis (programmed cell death) with 99mTc annexin V. J Nucl Med 1999;40: 184-191 64. Blankenberg FG, Strauss HW: Recent advances in the molecular imaging of programmed cell death: Part I—Pathophysiology and radiotracers. J Nucl Med 2012;534:1659-1662 65. Wang F, Fank W, Zhao M, et al: Imaging paclitaxel (chemotherapy)induced tumor apoptosis with 99mTc C2A, a domain of synaptotagmin I: A preliminary study. Nucl Med Biol 2008;35:359-364 66. Erba PA, Menfredi C, Lazzeri E, et al: Time course of paclitaxel-induced apoptosis in an experimental model of virus-induced breast cancer. J Nucl Med 2010;51:775-781 67. Watanabe A, Nishijima K, Zhao S, et al: Quantitative determination of apoptosis of pancreatic ß-cells in a murine model of Type 1 Diabetes Mellitus. J Nucl Med 2012;53:1585-1591 68. Deng Y, Chan SS, Chang S: Telomere dysfunction and tumour suppression: The senescence connection. Nat Rev Cancer 2008;8: 450-458 69. Xu L, Li S, Stohr BA: The role of telomere biology in cancer. Annu Rev Pathol Mech Dis 2013;8:49-78

C.R. Berry and P. Garg 70. Sharpless NE, DePinho RA: Telomeres, stem cells, senescence and cancer. J Clin Invest 2004;113:160-168 71. Liu M, Wang RF, Zhang CL, et al: Noninvasive imaging of human telomerase reverse transcriptase (hTERT) messenger RNA with 99mTcradiolabeled antisense probes in malignant tumors. J Nucl Med 2007;48: 2028-2036 72. Bergers G, Benjamin LE: Tumorigenesis and the angiogenic switch. Nat Rev Cancer 2003;3:401-410 73. Tlsty TD, Coussens LM: Tumor stroma and regulation of cancer development. Annu Rev Pathol Mech Dis 2006;1:119-150 74. Wang H, Zheng LF, Feng Y, et al: A comparison of 3D-CTA and 4D-CEMRA for the dynamic monitoring of angiogenesis in a rabbit VX2 tumor. Eur J Radiol 2012;81:104-110 75. Cai W, Chen X: Multimodality molecular imaging of tumor angiogenesis. J Nucl Med 2008;49:113S-128S 76. Hendrix MJC, Seftor EA, Hess AR, et al: Vasculogenic mimicry and tumour-cell plasticity: Lessons from melanoma. Nat Rev Cancer 2003;3:411-421 77. Carlin S, Paguachev A, Xiaorong S, et al: In vivo characterization of a reporter gene system for imaging hypoxia-induced gene expression. Nucl Med Biol 2009;36:821-831 78. Lin A, Hahn SM: Hypoxia imaging markers and applications for radiation treatment planning. Semin Nucl Med 2012;42:343-352 79. Lee TS, Scott AM: Hypoxia positron emission tomography imaging with 18-F-fluoromisonidazole. Semin Nucl Med 2007;37:451-461 80. Dewhirst MW, Cao Y, Moeller B: Hypoxia and metabolism: Cycling hypoxia and free radicals regulate angiogenesis and radiotherapy response. Nat Rev Cancer 2008;8:425-437 81. Carlin S, Humm JL: PET of hypoxia: Current and future perspectives. J Nucl Med 2012;53:1171-1174 82. Chitneni SK, Palmer GM, Zalutsky MR, et al: Molecular imaging of hypoxia. J Nucl Med 2011;52:165-168 83. Bavcar S, Argyle DJ: Receptor tyrosine kinase inhibitors: Molecularly targeted drugs for veterinary cancer therapy. Vet Comp Oncol 2012;10: 163-167 84. Mishani E, Hagooly A: Strategies for molecular imaging of epidermal growth factor receptor tyrosine kinase in cancer. J Nucl Med 2009;50: 1199-1202 85. Buchholz TA, Tu X, Ang KK, et al: Epidermal growth factor receptor expression correlates with poor survival in patients who have breast carcinoma treated with doxorubicin-based neoadjuvant chemotherapy. Cancer 2005;104:676-681 86. Ozao-Choy J, Ma G, Kao J, et al: The novel role of tyrosine kinase inhibitor in the reversal of immune suppression and modulation of tumor microenvironment for immune-based cancer therapies. Cancer Res 2009;69:2514-2522 87. Pan Q, Wei Bao L, Kleer CG, et al: Protein Kinase C is a predictive biomarker of aggressive breast cancer and a validated target for RNA interference anticancer therapy. Cancer Res 2005;65:8366-8371 88. Henley JD, Cummings OW, Loehrer PJ: Tyrosine kinase receptor expression in thymomas. J Cancer Res Clin Oncol 2004;130:222-224 89. Nagy JA, Dvorak AM, Dvorak HF: VEGF-α and the induction of pathological angiogenesis. Annu Rev Pathol Mech Dis 2007;2:251-275 90. Kalluri R: Basement membranes: Structure, assembly and role in tumor angiogenesis. Nat Rev Cancer 2003;3:422-433 91. Wang F, Niu G, Chen X, et al: Molecular imaging of microRNAs. Eur J Nucl Med Mol Imaging 2011;38:1572-1579 92. Hara T, Bansal A, DeGrado TR: Choline transporter as a novel target for molecular imaging of cancer. Mol Imaging 2006;5:498-509 93. Kwee SA, DeGrado TR, Talbot JN, et al: Cancer imaging with fluorine18-labeled choline derivatives. Semin Nucl Med 2007;37:420-428 94. DeGrado TR, Baldwin SW, Wang S, et al: Synthesis and evaluation of 18 F-labeled choline analogs as oncologic PET tracers. J Nucl Med 2001;42:1805-1814 95. Bhankoo KK, Williams SR, Florian CL, et al: Immortalization and transformation are associated with specific alterations in choline metabolism. Cancer Res 1996;56:4630-4635 96. Rajagopalan KN, Deberardinis RJ: Role of glutamine in cancer: Therapeutic and imaging implications. J Nucl Med 2011;52:1005-1008

Perspectives in molecular imaging 97. Qu W, Oya S, Lieberman BP, et al: Preparation and characterization of 11 L-[5- C]-glutamine for metabolic imaging of tumors. J Nucl Med 2012;53:98-105 98. van der Have F, Vastenhouw B, Ramakers RM, et al: U-SPECT-II: An ultra-high-resolution device for molecular small animal imaging. J Nucl Med 2009;50:599-605 99. Pichler BJ, Wehrl HF, Judenhofer MS: Latest advances in molecular imaging instrumentation. J Nucl Med 2008;49:5S-23S 100. Seiser EL, Thomas R, Richards KL, et al: Reading between the lines: Molecular characterization of five widely used canine lymphoid tumor cell lines. Vet Comp Oncol 2011;10:30-50 101. Yamada N, Mori T, Murakami M, et al: Fascin-1 expression in canine cutaneous and oral melanocytic tumors. Vet Comp Oncol 2011;10: 303-311 102. Nielson LN, McEvoy F, Jessen LR, et al: Investigation of a screening programme and the possible identification of biomarkers for earlydisseminated histiocytic sarcoma in Bernese Mountain dogs. Vet Comp Oncol 2011;10:124-134

75 103. Robben JH, Visser-Wisselaar HA, Rutteman GR, et al: In vitro and in vivo detection of functional somatostatin receptors in canine insulinomas. J Nucl Med 1997;38:1036-1042 104. Berry CR, DeGrado TR, Nutter F, et al: Imaging of pheochromocytoma in 2 dogs using p-[18F]fluorobenzylguanidine. Vet Radiol Ultrasound 2002;43:183-186 105. Vail DM, MacEwen EG: Spontaneously occurring tumors of companion animals as models for human cancer. Cancer Invest 2000;18:781-792 106. Basu S, Kwee TC, Gatenby R, et al: Evolving role of molecular imaging with PET in detecting and characterizing heterogeneity of cancer tissue at the primary and metastatic sites, a plausible explanation for failed attempts to cure malignant disorders. Eur J Nucl Med Mol Imaging 2011;38:987-991 107. Zalutsky MR, Reardon DA, Pozzi OR, et al: Targeted alpha-particle radiotherapy with 211At-labeled monoclonal antibodies. Nucl Med Biol 2007;34:779-785 108. Dadachova E: Cancer therapy with alpha-emitters labeled peptides. Semin Nucl Med 2010;40:204-208